BER Estimation for Linear Clock and Data Recovery Circuit

Dongwoo Hong*, Tim Kwang-Ting Cheng

*Corresponding author for this work

Research output: Chapter in Book/Conference Proceeding/ReportBook Chapterpeer-review

Abstract

BER is a widely used figure of merit to indicate the quality of a communication system. A BER at 10 − 12 or even lower is often required for most communication systems. To measure such a low BER (at the level of 10 − 12) with reasonable accuracy, it would be necessary to capture more than 1012 samples. As a result, the test time would be excessively long even if the system runs in the gigahertz range. Thus, the conventional testing method, which compares each captured bit for error detection, requires a long test time and is therefore not cost-effective for high-volume production testing [17]. One of the major contributors to the BER is jitter. Jitter is specified in terms of the RJ and the DJ in most high-speed communication standards [4, 18, 19]. These values are based on the time-domain characteristics (e.g. the peak-to-peak or the rms values), and the spectral contents of the jitter (e.g. the amplitude and the frequency of the PJ) cannot be extracted from them. The CDR circuit, which recovers the clock signal from the data, can tolerate a certain frequency range of jitter. Thus, both the jitter spectral information and the characteristics of the CDR circuit have to be considered for measuring the overall system performance (i.e. BER), if PJ is present.
Original languageEnglish
Title of host publicationLecture Notes in Electrical Engineering
PublisherSpringer
Pages19-40
ISBN (Electronic)18761119
ISBN (Print)9789048134427, 9789400730946, 9789048134434
DOIs
Publication statusPublished - 2010
Externally publishedYes

Fingerprint

Dive into the research topics of 'BER Estimation for Linear Clock and Data Recovery Circuit'. Together they form a unique fingerprint.

Cite this